DE102012009461A1 - A numerical control device comprising a machine time prediction unit and a machine error prediction unit - Google Patents

Abstract

In conjunction with a machine program which is machine-processed in a machining of a workpiece by a numerical control-controlled tool, interpolation data, a command position point sequence and a servo position point sequence for each processing span are simulated by designating speed data for indicating a machine speed and accuracy data Specifying a machine accuracy determined. A predicted machine time for the machining of the workpiece is determined based on the determined interpolation data, and a predicted machine error for the machining of the workpiece is determined based on the determined command and servo position point sequences. Furthermore, the accuracy data and the speed data for the shortest predicted machine time are determined within a preset machine error tolerance, based on a plurality of predicted machine times and a plurality of predicted machine errors.

Description

Background of the invention

Field of the invention

The present invention relates to a numerical control apparatus having a machine time prediction unit and a machine error prediction unit.

Description of the Related Art

In machine processing of a workpiece using a machine tool, in general, the accuracy of machine processing is reduced as the machine time is reduced or as the machine speed is increased. If the machine time is increased or the machine speed is reduced, on the other hand, the accuracy of the machine processing is improved. In the machine processing of the workpiece, the accuracy of the machine processing is more important than the machine time. Thus, users who use the machine tool to machine the tool want to be able to machine the workpiece in a minimum machine time with a precision of machine processing within a preset machine error tolerance. However, it is not easy to ensure the appropriate machine time and machine error level for the workpiece.

Accordingly, the machine processing is normally performed at a speed within an allowable accuracy range after a machine trial processing has been performed several times with a somewhat longer time. Before machine processing a workpiece, therefore, users want to know by simulation how accurately the workpiece will be machined with machine speed and machine conditions without machine trial processing. Furthermore, users desirably want to know by simulation the machine speed and machine conditions with which the workpiece can be machined with machine processing accuracy within a machine error tolerance in a minimum machine time without machine trial processing.

• (a) Die japanische Patentanmeldungsoffenlegung Nr. 2004-227028 (entsprechend der ungeprüften US-Patentanmeldung Nr. US 2005/0228533 A1) offenbart eine Technik, in der ein Maschinenprogramm per Simulation in Maschinenvorgänge unterteilt wird, eine Zykluszeit für jeden der unterteilten Maschinenvorgänge berechnet wird, und Informationen bezüglich des Ergebnisses der Simulation auf einer Anzeigeeinheit einer Endgeräteeinrichtung angezeigt werden. Des Weiteren offenbart die japanische Patentanmeldungsoffenlegung Nr. 2006-301440 eine Maschinenzeitberechnungseinrichtung, die eine genaue Achsenbewegungszeit selbst bei einer maschinellen Verarbeitung schnell berechnen kann, die eine numerische Steuereinrichtung mit einer Funktion zum optimalen Einstellen der Vorschubrate und der Beschleunigung/Abbremsung verwendet.

Basic prior art techniques relating to machine time prediction are disclosed in the following patents.

• (a) The Japanese Patent Application Laid-Open No. 2004-227028 (according to Unexamined US Patent Application No. US 2005/0228533 A1) discloses a technique in which a machine program is divided into machine operations by simulation, a cycle time is calculated for each of the divided machine operations, and information regarding the result of the simulation on a display unit a terminal device are displayed. Furthermore, the Japanese Patent Application Laid-Open No. 2006-301440 an engine time calculator that can quickly calculate an accurate axle travel time even in a machine processing using a numerical controller having a function of optimally adjusting the feed rate and the acceleration / deceleration.

• (b) Die japanische Patentanmeldungsoffenlegung Nr. 58-35607 (entsprechend dem US-Patent Nr. 4,543,625 ) offenbart eine numerische Steuervorrichtung, die in der Lage ist, einen Maschinenfehler beim Eckenschneiden und einen Kreisfehler im Kreisbogenschneiden auf Toleranzen einzuschränken.

Although machine time is predicted in the two patent documents described above, machine errors are not predicted. Therefore, these techniques can not satisfy the user demand to know by simulation the accuracy of the workpiece at the machine speed without a machine trial processing or by simulation to know the machine speed at which the workpiece with an accuracy within a machine error tolerance in a minimum machine time without a machine trial processing can be processed by machine.

• (b) The Japanese Patent Application Laid-Open No. 58-35607 (according to the U.S. Patent No. 4,543,625 ) discloses a numerical control apparatus capable of limiting a machine error in corner cutting and a circular error in circular arc cutting to tolerances.

• (c) Die japanische Patentanmeldungsoffenlegung Nr. 2011-60016 (entsprechend der ungeprüften US-Patentanmeldung Nr. US2011/0057599 A1) offenbart eine Bahnkurvenanzeigevorrichtung, die eine Funktion zum genauen Quantifizieren eines Fehlers einer dreidimensionalen Bahnkurve eines Maschinenwerkzeugs und zum Anzeigen oder Ausgeben des Fehlers aufweist.

In the technique disclosed in the above-described patent document, the engine error is calculated on the assumption that acceleration / deceleration with exponential characteristics is performed. However, in modern numerical control devices, complicated speed control is performed for smooth acceleration / deceleration characteristics, such as both linear and bell-shaped acceleration / deceleration characteristics, and exponential acceleration / deceleration characteristics. Thus, the machine error can not be predicted by a simple computerized formula described in this patent document.

• (c) The Japanese Patent Application Laid-Open No. 2011-60016 (corresponding to Unexamined US Patent Application No. US2011 / 0057599 A1) discloses a trajectory display device which has a function of accurately quantifying a fault of a three-dimensional trajectory of a machine tool and displaying or outputting the error.

The trajectory display device disclosed in the above-described patent document comprises a command line segment definition unit and an error calculation unit. The command line segment definition unit defines command line segments that each connect two adjacent points for each command position point. The error calculation unit calculates, as an error of an actual position at each time point relative to a command path, the length of the shortest of vertical lines individually drawn from the actual position to the command line segments, or the length of a line segment representing the actual line position. Position and the closest command position, whichever is the shorter, connects. Data regarding the actual position are determined by actual operation of a drive shaft (servo), not by simulation.

The technique disclosed in the above-described patent document will be described below with reference to FIG 1 described. In this technique, a length from an actual position Q _m '' (m = 1, 2, ...) to a command position P _n (n = 1, 2, ...) is regarded as an error. In another case, on the other hand, a length from the command position P _n to the actual position Q _m "should be regarded as an error. In machine processing, for example, a right-angled corner by the technique disclosed in this patent document, as in 1 shown, lengths from actual positions to command positions (indicated by dashed line) are considered as errors. In this case, however, the length of a vertical line (indicated by a dashed-dotted line) is larger, which is drawn from a command position P ₄ at the corner to a line segment connecting actual positions Q ₄ "and Q ₅ ^" . Therefore, the length of this vertical line should be considered a mistake. Thus, there is also a case where the length from the command position to the actual position is to be regarded as an error, so that the method of error evaluation disclosed in this patent document is unsatisfactory.

Further, according to the technique disclosed in this patent document, the error is merely calculated and displayed or output. Therefore, this technique can not satisfy the user demand to know by simulation the accuracy of the workpiece with the machine speed and the machine conditions without a machine trial processing, or by simulation to determine the machine speed and machine conditions with which the workpiece with an accuracy within a machine error tolerance in one machine time can be machine-processed without a machine test-run processing.

Usually, a numerical control device has a machine condition selection function. According to this function, engine condition data including an allowable acceleration and a permissible corner speed difference are calculated in 2 are set in advance as parameters, and the accuracy data to be used for the machining will be expressed as Rr (r = 1 to 10) in the block G05.1 in the 3 shown program.

According to the in 2 As shown in FIG. 14, the engine condition data (including allowable acceleration and allowable corner speed difference) are previously set in advance for each of the speed-oriented data (precision data 1) and accuracy-oriented data (precision data 10), and machine condition data for accuracy data 2 to 9 is interposed between the accuracy data 1 and 10 determined by a proportional distribution based on those set values. Although only the two typical engine condition data are described herein, various other engine condition data such as a jump (acceleration variation), a post-interpolation acceleration / deceleration time constant, etc. may be set in the same manner.

3 shows instructions for specifying accuracy data to be displayed among the accuracy data described above.

"G05.1Q1" in the first block represents a block for a machine condition selection command, and "r" and "Rr" indicate the accuracy data consisting of the in 2 shown accuracy data 1 to 10 are selected. For example, "r" is set to a small value for coarse machine processing or a large value for fine machine processing. It should be understood that here, the machine condition data including the allowable acceleration, the allowable corner speed difference, etc., may be directly commanded or set parameter values for these condition data instead of the accuracy data using the numbers 1 to 10.

While the F command in the second block is a speed command as a machine command, an actual command speed may be changed to a value (F command) × (override value) using an override (1 to 200%). Furthermore, a parameter-substituted Speed can be used for a command speed that ignores the F command. Furthermore, the F command in the program can be changed to a modified speed command.

Will "G05. 1Q1R4 "in the in 2 For example, when instructed as shown in FIG. 14, then machining with the allowable acceleration of 1,533 (= (6 × 2,000 + 3 × 600) / 9) mm / s ² and allowable corner speed difference of 800 (= (6 × 1,000 + 3 × 600) / 9) mm / min, which are set as parameters for the machine condition data. Alternatively, as described above, these data may be instructed directly, or corresponding parameter set values may be set. Although the accuracy data can be easily instructed by this engine condition selection function, however, the proper engine failure level and the engine time for all the accuracy data can not be ensured.

Summary of the invention

Accordingly, in view of the above-described problems of the prior art, an object of the invention is to provide a numerical control apparatus comprising: a machine time prediction unit that predicts a predicted machine time by simulation and a machine error prediction unit that simulates a predicted machine error; Accuracy data and speed data are displayed as machine time / machine error prediction data and a machine program is designated. Another object of the invention is to provide a numerical control apparatus in which a plurality of predicted machine times and a plurality of predicted machine errors are determined on the basis of the designated accuracy data and speed data, and the shortest predicted machine time within a preset engine error tolerance based thereon predicted machine times and the predicted machine error.

A numerical control device according to the invention controls a machine tool, which processes a workpiece on the basis of a machine program. The numerical control apparatus comprises an expelling unit for displaying speed data for outputting a machine speed and accuracy data for outputting an engine accuracy, a simulation unit for simulation-defining an instruction position point sequence and a servo position point sequence for each processing span, using the speed data and the accuracy data indicative of the engine program by the designation unit, a machine time prediction unit that determines a predicted machine time during which the workpiece is machined in the simulation unit, and a machine error prediction unit that determines a predicted machine error caused when the workpiece using the instruction position point sequence and the servo position point sequence is processed by the simulation unit.

The machine error prediction unit may acquire the instruction position point sequence and the servo position point sequence for each processing period of simulation by the simulation unit, a distance from a specific instruction position in the instruction position point sequence to the servo position point sequence or a right angle line drawn from the specific instruction position to a line segment may be the adjacent points in any servo position, whichever is shorter than an error in the specific instruction position, and the largest of the errors in the instruction positions can be determined as an error between the instruction position point sequence and the servo position point sequence.

The engine error prediction unit may obtain the command position point sequence and the servo position point sequence for each processing span of simulation by the simulation unit, may include a distance from a specific servo position in the servo position point sequence to the command position point sequence or a vertical line drawn from the specific servo position to a line segment, the adjacent points in each instruction position, whichever is shorter than an error in the specific servo position, and the largest of the errors in the servo positions can be determined as an error between the instruction position point sequence and the servo position point sequence.

The engine error prediction unit may acquire the command position point sequence and the servo position point sequence for each processing span of simulation by the simulation unit, may include a distance from a specific command position in the command position point sequence to the servo position point sequence or a perpendicular line drawn from the specific command position to a line segment having adjacent points in connecting the servo position, whichever is the shorter, than determining an error in the specific instruction position, the largest of the errors in the instruction positions may be as an error may be determined from the command position point sequence to the servo position point sequence, a distance from a specific servo position in the servo position point sequence to the command position point sequence or a perpendicular line drawn from the specific servo position to a line segment connecting adjacent points in each command position may as well always being shorter than an error in the specific servo position is determined, the largest of the errors in the servo positions may be determined as an error from the servo position sequence to the instruction position point sequence, and the greater of the error from the instruction position point sequence to the servo position point sequence or error from the servo position sequence to the instruction position point sequence are determined as an error between the instruction position point sequence and the servo position point sequence.

The machine time prediction unit may determine a plurality of predicted machine times based on a plurality of designated accuracy data, and the engine error prediction unit may determine a plurality of predicted engine errors based on a plurality of designated speed data. The accuracy data and the speed data for the shortest predicted engine time within a default engine error tolerance may be determined based on the plurality of engine predicted times and the plurality of engine predicted errors thus determined.

The numerical control device may control a multi-axis machine having a rotation axis, the machine error prediction unit may determine a three-dimensional command position sequence and a three-dimensional servo position point sequence of a tool center point position based on the command position point sequence and the servo position point sequence, and the error between the command position point sequence and the servo position point sequence with the three-dimensional command position point sequence and the three-dimensional Servo position sequence, which are respectively used as the instruction position point sequence and the servo position point sequence.

The accuracy data may be given by the engine condition data based on an allowable acceleration or a permissible corner speed difference.

According to the invention, there can be provided a numerical control apparatus comprising a machine time prediction unit that predicts a predicted machine time by simulation and a machine error prediction unit that predicts a predicted machine error, wherein accuracy data and speed data are designated as machine time / machine error prediction data and a machine program becomes. In the numerical control apparatus, a plurality of predicted machine times and a plurality of predicted machine errors are determined on the basis of these designated accuracy data and speed data, and the shortest predicted machine time becomes within a preset machine error tolerance based on a plurality of predicted machine times and a plurality of predicted machine errors certainly.

Brief description of the drawings

The above-described and other objects and features of the invention will become apparent from the following description of embodiments with reference to the accompanying drawings. Show it:

1 Fig. 10 is a diagram for describing a technique relating to an error of an actual position relative to a command path disclosed in a prior art document;

2 a diagram describing a machine condition selection function according to the prior art;

3 a diagram describing program instructions including a command for selecting accuracy data to be designated;

4 a diagram describing features of the present invention;

5 5 is a functional block diagram of a numerical control device including a simulation unit, a machine time prediction unit, and a machine error predicting unit for a machine program according to the present invention;

6 a diagram describing a servo simulation unit for simulating a servo operation;

7 a diagram describing a machine time for a block in question;

8th a diagram describing interpolation data for each processing span;

9 a flowchart describing processing in the machine error prediction unit;

10 a diagram illustrating the process of step SA06 in the flowchart according to 9 describes;

11 a flowchart illustrating the process of step SA03 in the flowchart according to 9 describes in more detail;

12 a diagram describing how to determine an error of a point sequence P _n ;

13 an example of a predicted machine time and a predicted machine error;

14 a table describing predicted machine times and predicted machine errors determined based on accuracy data and speed data;

15 a schematic view of a five-axis machine with a rotatable tool head;

16 a diagram describing relationships between P _n and Q _m and between P _n 'and Q _m ' in the five-axis machine; and

17 12 is a block diagram schematically describing a structure of an embodiment of the numerical control apparatus including the machine time prediction unit and the machine error prediction unit according to the present invention.

Detailed Description of the Preferred Embodiments

First, features of the present invention will be described with reference to FIG 4 to be discribed.

The present invention is intended to simulate machine and machine timing predictions for one or more sets of speed data and accuracy data, and to estimate machine processing accuracies (machine errors) and machine times for workpieces calculated according to given speed data and accuracy data (see (1) according to 4 ) and machine speeds (speed data) and machine conditions (accuracy data) are to be machine processed to machine workpieces with accuracies within a machine error tolerance in a shortest machine time (see (2) 4 ).

According to the present invention, errors to be evaluated include simulated errors from servo positions to instruction positions and from instruction positions to servo positions. In this way, the errors can be predicted more accurately.

Under these circumstances, the predicted machine errors and times can be determined by simulation before the actual machining, based on the estimation of the speed data and accuracy data ((1) according to FIG 4 ), so that production planning can be made easier. Further, a machine tool can be operated more efficiently by minimizing the machine time, based on the estimation of the engine speeds (speed data) and the engine conditions (accuracy data) ((2) of FIG 4 ). Consequently, further, an effect can be obtained that energy-saving machining can be achieved.

Hereinafter, a description of the embodiments 1 to 3 is given.

[Embodiment 1]

5 shows a functional block diagram of a numerical control device 1 according to the present invention, comprising a machine time prediction unit and a machine error prediction unit for a machine program.

A program analysis unit 3 simulates a program analysis process for reading and analyzing a machine program 2 and for creating interpolation data. An interpolation unit 4 simulates an interpolation process so that interpolation is performed according to the interpolation data, and interpolation data ΔP _{n is} established based on velocity data for blocks and vertices between blocks by a pre-interpolation acceleration / deceleration unit 5 were created. A post interpolation acceleration / deceleration unit 7 simulates a post-interpolation acceleration / deceleration operation so that post-interpolation acceleration / deceleration is performed for the interpolation data ΔP _n , and servo position command data VC _{n is created} by accumulating these data. The servo position command data VC _n are from the post-interpolation acceleration / deceleration unit 7 a servo simulation unit 8th supplied, the servo position data Q _n (hereinafter simply referred to as the servo position Q _n ) as a simulation of an actual servo operation based on the servo position command data VC _n created. A simulation unit includes the program analysis unit, the interpolation unit, the post-interpolation acceleration / deceleration unit, the servo simulation unit, and a machine time prediction unit (described below) 6 ,

The servo simulation unit 8th for simulating a servo operation will be described below with reference to FIG 6 to be discribed.

The servo position command data VC _n are from the post-interpolation acceleration / deceleration unit 7 in the servo simulation unit 8th entered with each processing span. The servo position command data VC _n entering the servo simulation unit 8th are entered into a forward simulation unit 80 entered. The forward simulation unit 80 differentiates the servo position command data VC _n and outputs the result of multiplication by a forward coefficient as ΔU _n .

The difference between the servo position command data VC _n and a servo position (feedback data) Q _n-1 for the subsequent processing period becomes a position loop increment simulation unit 81 entered. The simulation unit 81 outputs ΔR _n obtained by multiplying the input difference by a position loop gain. Then, the sum of ΔU _n and ΔR _n becomes a position simulation unit 82 is entered, and the accumulation of the sum is obtained as the servo position Q _n .

Data with the index " _n " or another index " _m " (described below) is a vector with as many data entries as feed axes.

The operations described above are performed with reference to various machine time / machine error prediction data, such as parameters, macro variables, DI signals, etc. Further, in these operations, servo operation is not actually performed, but all are simulated.

The parameters, macro variables, DI signals, etc. of the machine time / machine error prediction data are formed in addition to machine data of the same kind. In this case, the DI signals are therefore imaginary signal data for machine time prediction and machine error prediction.

As described above, the servo simulation unit performs 8th Simulations of a delay based on the position loop gain and a servo operation based on a feedforward control. Thus, the servo simulation unit becomes 8th in addition to a servo processing unit for machine processing. The program analysis unit 3 , the interpolation unit 4 , the pre-interpolation acceleration / deceleration unit 5 and the post-interpolation acceleration / deceleration unit 7 may be provided in addition to processing units of the same types for machine processing. Alternatively, the processing units of the same types of machine processing (foreground) may be used in the background. The structure of the servo simulation unit 8th according to 6 is merely exemplified, and a servo simulation may be performed by a servo simulation unit of another construction.

Simulations for the program analysis unit 3 , the interpolation unit 4 , the pre-interpolation acceleration / deceleration unit 5 and the post-interpolation acceleration / deceleration unit 7 are processed by a conventional method which will not be described in detail. However, the processing spans for the interpolation unit may be 4 , the pre-interpolation acceleration / deceleration unit 5 and the post-interpolation acceleration / deceleration unit 7 for the simulation to be longer than those for the machine processing. For example, even if the processing spans of the interpolation unit, the pre-interpolation acceleration / deceleration unit, and the post-interpolation acceleration / deceleration unit for machining are one millisecond, those for the interpolation unit 4 , the pre-interpolation acceleration / deceleration unit 5 and the post-interpolation acceleration / deceleration unit 7 for machine time prediction and machine error prediction for the simulation to be assumed to be up to about 8 ms. In this way, the machine time prediction and the machine error prediction can be performed faster. This is because a coarser calculation can be made for 8 ms than that for 1 ms fast for the simulation.

The machine time prediction unit 6 determines a predicted machine time Te based on the interpolation data created by the interpolation unit. For example, the calculation for the machine time prediction may be based on the form of the interpolation data, or the machine time may be predicted by counting the number of interpolations. In the case of linear acceleration / deceleration, for example, for the interpolation data shape of each block for the computation, an upper limit is defined by the command speed of the block in question and accelerations of the block in question by the pre-interpolation acceleration / deceleration unit 5 are created, and are indicated by solid lines, as in 7 shown. Further, trough speeds are regarded as corner speeds for the block created by the pre-interpolation acceleration / deceleration unit and the area as a command block length. Thus, the predicted machine time for the block in question can be calculated.

Specifically, all the interpolation data ΔP _{n have} a rectangular shape for each processing span, as in FIG 8th shown. In general, however, the processing time of the block in question is sufficiently longer than the processing margin, so that 7 shows an entire shape that includes the individual rectangular shapes that are joined together. The prediction by counting the number of interpolations is a method in which the engine time is calculated by multiplying the frequency (number of interpolation data ΔP _n ) by the generation of the interpolation data ΔP _n from the interpolation unit 4 is determined with the processing margin. Alternatively, the engine time may be reported by multiplying the count number of the other data created in the simulation unit, such as the servo position command data VC _n created by the post-interpolation acceleration / deceleration unit, the servo position Q _n created by the servo simulation unit, and so forth the processing margin. Since these machine time prediction methods are conventional technologies, they will not be described in detail below.

A machine error prediction unit 9 , as an error, predicts the difference between an interpolation path based on the interpolation data ΔP _n and a deceleration path based on the post-interpolation acceleration / deceleration and the servo operation simulation. As described above, the engine error predicting unit receives 9 the servo position Q _n of the servo simulation unit 8th and also receives the interpolation data ΔP _n as output data from the interpolation unit 4 , The machine error prediction unit 9 determines an instruction position P _n = P _n -1 + ΔP _n by accumulation. In this way, a command position point sequence P _n and a servo position point sequence Q _{n are} created. The dot sequence Q _n includes, by the number of processing intervals for the delay by the post interpolation acceleration / deceleration and servo operation simulation, more points than the dot sequence P _n . In the following description, therefore, the alternative index " _m " is used to obtain Q _m (m = 0, 1, 2, ..., me) for P _n (n = 0, 1, 2, ..., ne) display. The symbols ne and me show the numbers for the individual final points in the point sequences. In general, there is a relationship in which ne <me.

The processing of the machine error prediction unit 9 will be described below with reference to the flowchart according to 9 being represented.

[Step SA01] Conditions n = 0, m = 0, maxep = 0, and maxep = 0 are specified.

[Step SA02] The point sequence P _n is formed for (n-dn ≦ _n ≦ + dn) according to a number dn previously set to define its range. If a range (0 ≦ n ≦ ne) is exceeded by (n-dn ≦ n ≦ + dn), then only the dot sequences within the range (0 ≦ n ≦ ne) are provided. Similarly, the dot sequence Q _{m is formed} for (m-dm ≦ _m ≦ m + dm) according to a number dm that is pre-set to define its range. If a range (0 ≦ m ≦ me) is exceeded by (m-dm ≦ m ≦ m + dm), then only the dot sequences within the range (0 ≦ m ≦ ne) are provided.

[Step SA03] Error calculations are performed from a specific command position P _n in the command position point sequence P _n to the point sequence Q _m to obtain the errors Ep from the specific command position P _n to the point sequence Q _m . Furthermore, the maximum value Maxep of the error Ep and m is updated. This step will be described in detail below with reference to the flowchart of FIG 11 to be discribed.

[Step SA04] Error calculations are performed from a specific servo position Q _m in the servo position point sequence Q _m to the point sequence P _n to obtain the errors Eq from the specific servo position Q _m to the point sequence P _n . Furthermore, the maximum value Maxeq of the error Eq and n is updated.

[Step SA05] It is determined whether or not n = ne and m = me are reached. If the decision is No, then the program returns to step SA02. If the decision is yes, the program proceeds to step SA06.

[Step SA06] The maximum value Maxep or Maxeq, whichever is larger, is determined as a predicted machine error Ee between the command position point sequence P _n and the servo position point sequence Q _m , whereupon the processing ends.

More specifically, regarding the command position point sequence P _n and the servo position point sequence Q _m, the errors Ep from the specific command position P _n to the point sequence Q _m and the error Eq from the specific servo position Q _m to the point sequence P _n for the regions (n-dn ≦ n ≤ + dn) and (m-dm ≦ m ≦ m + dm). The numbers n and m are updated when this calculation is performed. The maximum value Maxep (= Max (Ep)) of the error Ep and the maximum value Maxeq (= Max (Eq)) of the error Eq are obtained, and the maximum value Maxep or Maxeq, whichever is larger, becomes the predicted machine error Ee (= Max (Maxep, Maxeq)) between the command position point sequence P _n and the servo position point sequence Q _m . Thus, Max (Ep) represents the maximum of a plurality of errors Ep, Max (Eq) represents the maximum of a plurality of errors Eq, and Max (Maxep, Maxeq) represents the larger of the maximum values Maxep and Maxeq, as in 10 shown.

In step SA03, error calculations are performed from a specific command position P _n to the point sequence Q _m , and the maximum value Maxep of the error Ep is established. Similarly, in step SA04, error calculations are performed from a specific servo position Q _m to the point sequence P _n , and the maximum value Maxeq of the error Eq is established. Only one of steps SA03 or SA04 may be performed instead of performing both of these steps.

In the technique described above, described in the Japanese Patent Application Disclosure No. 2011-60016 is disclosed, only the calculation corresponding to step SA04 is performed. However, as described above, it should be noted that in the case of the patent document, the calculation is performed for the actual position obtained by actually operating a drive shaft (servo), whereas in the case of the present invention, the calculation for the position is by simulation is carried out.

It has been described above that an error is calculated after the completed command position point sequence P _n (n = 0, 1, 2, ..., ne) and the servo position point sequence Q _m (m = 0, 1, 2,..., Me ) are created. Alternatively, however, a similar calculation can be performed by obtaining the necessary command position P _n (n - dn ≤ n ≤ + dn) and the servo position Q _m (m - dm ≤ m ≤ m + dm), while the machine program block by block is read as required , In this case, the entire point sequences P _n and Q _m need not be prepared in advance, so that the memory capacity can be saved.

The command position P _n, the specific command position P _n, the position command point sequence P _n and P _n point sequence, all of which are indicated by the symbol P _n, are expressions that have slightly different meanings. The command position P _n represents a single command position. The specific command position P _n represents a specific or noticeable command position. The command position point sequence P _n and both the point sequence P _n represent a command point sequence P _n represents that includes a plurality of command positions. This also applies to the case of Q _m .

Step SA03 in the flowchart of FIG 9 will be described below with reference to the flowchart according to 11 to be discribed.

[Step SB01] Initial values of i, im and EP are written as i = m-dm, im = i and EP = | P _n -Q _i | set. With respect to the value i, i = m-dm is generally indicated, but the range of (0 ≤ m ≤ me) is exceeded by (m-dm ≤ m ≤ + dm) for Q _m as described above in connection with the description of step SA02 in the flowchart of FIG 9 however, i is set to the smallest value within the range in which (m-dm ≦ m ≦ + dm) and (0 ≦ m ≦ me) overlap each other.

[Step SB02] A length | P _n - Q _{i + 1} | between P _n and Q _{i + 1} is calculated. Is | P _n - Q _{i + 1} | less than Ep, then Ep = | P _n - Q _{i + 1} | and im = i + 1, and then the value i + 1 is stored.

[Step SB03] A line segment Lq connecting Q _i and Q _{i + 1} is created.

[Step SB04] It is determined whether or not a perpendicular line from P _n to the line segment Lq can be drawn. If the vertical line can not be drawn, the program proceeds to step SB07. If the vertical line can be drawn, the program proceeds to step SB05.

[Step SB05] A length Lper of the vertical line drawn from P _n to the line segment Lq is calculated.

[Step SB06] If Lper is smaller than Ep, then Ep = Lper and im = i + 1 are set, and then the value i + 1 is stored.

[Step SB07] It is determined whether or not i = m + dm-1 is reached. If the decision is yes, the program proceeds to step SB09. If the decision is No, then the program returns to step SB08. Here, it is generally determined whether or not i = m + dm-1 is achieved. When the range (0 ≦ m ≦ me) is exceeded by (m-dm ≦ m ≦ + dm) for Q _m as described above in connection with the description of step SA02 in the flowchart of FIG 9 however, it is determined whether or not i is equal to or smaller than the largest value within one range by one (m - dm ≦ m ≦ + dm) and (0 ≤ m ≤ me) overlap each other. Thus, the shorter one of the vertical lines drawn from P _n to line segments connecting adjacent points in the point sequence Q _m (m-dm ≦ _m ≦ + dm) can be obtained as Ep. Thus, as in 12 The value Ep is set as an error between the value P _n and the point sequence Q _m (m-dm ≦ _m ≦ + dm).

[Step SB08] 1 is added to i to update i, whereupon the program returns to step SB02 and the processing is continued.

[Step SB09] If Ep is larger than Maxep, then Maxep = Ep is set, updating the maximum value Maxep of Ep.

[Step SB10] If greater than m, then m = im is set. If not greater than m, then m = m + 1 is set to update m. However, if m exceeds the maximum value me, m is regarded as equal to me, whereupon the processing ends.

The predicted machine time Te during which the workpiece will be machined and the predicted machine error Ee that will be caused when the workpiece is machined may be determined if the machine program is identified and executed by outputting the accuracy data and speed data; which in conjunction with the machine condition selection function (cf. 2 ) are described to the numerical control device according to 5 indicating the machine time prediction unit 6 and the engine error prediction unit 9 includes. The speed data may be output by various methods, for example, by applying an override as a DI signal, applying the parameter set speed based on a parameter, or by changing the F command in the program to a modified speed command.

It should be understood that only a portion of the machine program may be considered to be an object of machine time prediction and machine error prediction. For example, only a mode part of a slice command (G01, G02, G03, etc.) may be used as an object to be predicted. Alternatively, only a block of the machine program to be cut by a particular tool or only a specific process part may be used as an object to be predicted.

13 FIG. 15 shows an example in which the predicted engine time Te and the predicted engine failure Ee are obtained with the precision data at 9 and the speed data (override) at 50%. The predicted machine time Te (minutes) or the predicted machine error Ee (mm) are entered in the upper and lower columns, respectively. Based on the precision data and the speed data (override) thus designated, the predicted machine time and the predicted machine error can be simultaneously determined by simulation.

[Embodiment 2]

In Embodiment 2, conditions for the shortest machine time within the machine error tolerance are determined.

An in 14 The table shown is made up of predicted machine times Te and predicted machine errors Ee determined based on some of the converted accuracy data and speed data.

In this case, parameter-set speeds are given as the velocity data. 14 shows predicted machine times and predicted machine errors represented by those data in five rimmed areas, ie (precision data: 1, speed data: 1,000 mm / min), (prediction data 10, speed data 1,000 mm / min), (accuracy data: 5, speed data: 2,000 mm / min), (Accuracy data: 1, Velocity data: 3,000 mm / min) and (Accuracy data: 10, Velocity data: 3,000 mm / min). The predicted machine time Te (minutes) and the predicted machine error Ee (mm) are respectively plotted in the upper and lower columns of each framed area. Data in the other areas is obtained by interpolating the bordered data by proportional distribution. Of course, data can be simulated in some areas that are different from the five outlined areas, or in all areas.

If the default machine error tolerance is 0.05 mm then data in painted areas will be excluded as they exceed the tolerance. Thus, data for the shortest predicted machine time within the machine error tolerance can be easily determined as (accuracy data: 9, speed data: 3,000 mm / min).

[Embodiment 3]

In the case of a multi-axis machine having axes of rotation (X, Y, Z, B, and C axes) in a tool head and tables, a machine error prediction unit 9 a three-dimensional command position sequence P _n '(P _Xn ', P _Yn ', P _Zn ') and determine a three-dimensional servo position point sequence Q _m '(Q _Xm ', Q _Ym ', Q _Zm ') of a tool center point position based on instruction position point sequences P _n (P _Xn , P _Yn , P _Zn , P _Bn , P _Cn ) and servo position point sequence Q _m (Q _Xm , Q _Ym , Q _Zm , Q _Bm , Q _Cm ) and a tool length, and predict a machine error by managing the particular instruction position point sequences and servo position point sequences in the same manner as the instruction position point sequences and servo position point sequences according to the first or second embodiment.

Although indices B and C are used to represent positions of the assumed B and C axes of the multi-axis machine, the machine may have other axes of rotation, A and B axes or A and C axes. Furthermore, multi-axis machines include a five-axis machine with a rotatable tool head, a multi-axis machine with a rotatable table, and a five-axis machine of the compound type (with a rotatable tool head and a rotatable table). In each of these machines, the tool center point position is determined by the three-dimensional command position point sequences P _n '(P _Xn ', P _Yn ', P _Zn ') and the three-dimensional servo position point sequences Q _m '(Q _Xm ', Q _Ym ', Q _Zm ') of a tool center point position a coordinate system of a table indicated on which a machined workpiece is placed. In the case of the multi-axis machine with a rotatable table, the coordinate system of a table on which a workpiece to be machined is placed is a coordinate system rotating on and together with the rotating table.

15 shows a schematic view of a tool head portion of a five-axis machine with a rotatable tool head, and 16 shows a diagram describing relationships between P _n and Q _m and between P _n 'and Q _m ' in the five-axis machine.

17 FIG. 12 is a block diagram schematically describing a structure of an embodiment of a numerical control apparatus including the machine time prediction unit and the machine error prediction unit according to the present invention. FIG.

The numerical control device 1 has the machine condition selection function described with reference to FIG 2 is described. A CPU 111 for use as a processor controls the entire numerical control device 1 according to a system program stored in a ROM 112 is stored. A RAM 113 is loaded with various data or input or output signals. Different data stored in a non-volatile memory 114 are stored in their current state, even after a power cut. A graphics control circuit 115 converts digital signals to display signals and feeds them to a display device 116 to. A keyboard 117 is a device with numeric keys and character keys for entering various setting data.

An axis control circuit 118 receives motion commands for each axis from the CPU 111 and gives this to a servo amplifier 119 out. Upon receipt of the motion commands, the amplifier controls 119 Servo motors of a machine tool (not shown) 20 at. These constituent elements are connected to each other by a bus 121 connected. A programmable machine controller (PMC) 122 receives a T-function signal (tool selection command) and the like through the bus 121 during the execution of the machine program. Then the PMC processes 122 this signal according to a sequence program and outputs this as an operation command, whereby the machine tool 20 is controlled. Upon receipt of a status signal from the machine tool 20 transfers the PMC 122 Furthermore, a required input signal to the CPU 111 , Furthermore, the bus 121 with a software key 123 whose function changes according to the system program or the like, and an interface 124 by which NC data is supplied to an external device, such as a memory device. The software button 123 , together with the display device 116 and the keyboard 117 , is on a display / manual data entry (MDI) panel 125 created.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• JP 2004-227028 [0004]
• JP 2006-301440 [0004]
• JP 58-35607 [0005]
• US 4543625 [0005]
• JP 2011-60016 [0006, 0070]

Claims (7)

A numerical control apparatus for drivingly controlling a machine tool that machines a workpiece based on a machine program, the numerical control apparatus comprising: an expelling unit for displaying speed data for indicating an engine speed and accuracy data for indicating an engine accuracy; a simulation unit that determines an instruction position point sequence and a servo position point sequence for each processing span by simulating a program analysis operation, an interpolation operation, a post interpolation acceleration / deceleration operation, and a servo operation using the speed data and the accuracy data designated for the engine program by the designation unit; a machine time prediction unit that determines a predicted machine time during which the workpiece will be machined, in the simulation unit; and a machine error predicting unit that determines a predicted machine error caused when the workpiece will be machined using the command position point sequence and the servo position point sequence determined by the simulation unit. A numerical control apparatus according to claim 1, wherein the engine error prediction unit obtains the command position sequence and the servo position sequence for each processing period of the simulation by the simulation unit, a distance from a specific command position in the command position point sequence to the servo position point sequence or a perpendicular line drawn from the specific command position to a line segment connecting adjacent points in each servo position whichever is shorter than an error in the specific command position will, and the largest of the errors in the command positions is determined as an error between the command position sequence and the servo position sequence. A numerical control apparatus according to claim 1, wherein the engine error prediction unit obtains the command position sequence and the servo position sequence for each processing period of the simulation by the simulation unit, a distance from a specific servo position in the servo position point sequence to the instruction position point sequence or a vertical line drawn from the specific servo position to a line segment connecting adjacent points in each instruction position, whichever is the shorter, than an error in the specific servo position is determined, and the largest of the errors in the servo positions is determined as an error between the command position sequence and the servo position sequence. A numerical control apparatus according to claim 1, wherein the engine error prediction unit obtains the command position sequence and the servo position sequence for each processing period of the simulation by the simulation unit, a distance from a specific command position in the command position point sequence to the servo position point sequence or a perpendicular line drawn from the specific command position to a line segment connecting adjacent points in each servo position whichever is shorter than an error in the specific command position becomes, the largest of the errors in the command positions is determined as an error from the command position sequence to the servo position sequence, a distance from a specific servo position in the servo position point sequence to the instruction position point sequence or a vertical line drawn from the specific servo position to a line segment connecting adjacent points in each instruction position whichever is shorter than an error in the specific servo position becomes, the largest of the errors in the servo positions is determined as an error from the servo position sequence to the instruction position point sequence, and the greater of the error from the command position sequence to the servo position sequence and the error from the servo position sequence to the command position point sequence is determined as an error between the command position point sequence and the servo position point sequence. The numerical control apparatus according to claim 1, wherein the engine time prediction unit and the engine error prediction unit determine a plurality of predicted engine times and a plurality of predicted engine errors based on a plurality of designated accuracy data and a plurality of designated speed data, and the accuracy data and the speed data for the shortest predicted engine time within a preset machine error tolerance, based on the plurality of predicted machine times determined thereby and the plurality of predicted machine errors. A numerical control apparatus according to claim 1, wherein the numerical control device controls a multi-axis machine having a rotation axis, the engine error prediction unit determines a three-dimensional command position sequence and a three-dimensional servo position point sequence of a tool center point position based on the command position point sequence and the servo position point sequence, and the error between the command position sequence and the servo position point sequence is determined with the three-dimensional command position sequence and the three-dimensional servo position point sequence respectively used as the command position point sequence and the servo position point sequence. The numerical control apparatus according to claim 1, wherein the accuracy data is indicated by machine condition data based on an allowable acceleration or an allowable corner speed difference.

Images